[EM] Does MAM use the Copeland method?

[Steve Eppley]

Hi,

Paul K asked:
> Why should the 33 percent of the voting population who 
> most dislike "A" be the cause of A to win, whereas 56
> percent rank C>A ? 

If the "B>C>A" votes of those 33% are deleted, then 
A still wins.  I don't yet understand Paul's concern.

Paul, what's your favorite voting method?  How does 
it tally that example, which I repeat here:

    The votes:
    4: A>B>C
    3: B>C>A
    2: C>A>B

Most methods elect A, as MAM does.  Plurality elects A, 
IRV elects A, etc.  Borda returns a tie between A & B, 
Copeland returns a 3-way tie.  So I'm curious to learn
why Paul thinks A should be defeated...

If Paul thinks C should be elected, then I will point 
out that a majority larger than 56% ranked B over C, 
and if he thinks B should be elected, I will point out
that a majority larger than 56% ranked A over B.

Furthermore, the finish order A>B>C is the only ordering
that satisfies the specs in the Immunity from Majority 
Complaints criterion, and MAM is the only method that 
is always immune from majority complaints. (I refer to 
the criterion defined in my web pages, not the weaker 
criterion that Jobst wrote and to which he unfortunately 
gave the same name, which was forwarded to this maillist 
a couple of weeks ago in a message about Jobst's River
method.  Forwarded by Ralph, if memory serves.)

--Steve